Child Abuse & Neglect 64 (2017) 89–100

Contents lists available at ScienceDirect

Child Abuse & Neglect

Childhood adversity and adult depression: The protective

role of psychological resilience
Julia C. Poole a,∗ , Keith S. Dobson a , Dennis Pusch b
a
Depression Research Laboratory, University of Calgary, 2500 University Drive NW, Calgary, Alberta T2N 1N4, Canada
b
Alberta Health Services, Suite 1150, 10201 Southport Road SW, Calgary, Alberta T2W 4X9, Canada

a r t i c l e i n f o a b s t r a c t

Article history: Adverse childhood experiences (ACEs), such as childhood abuse, neglect, and household
Received 6 September 2016 dysfunction, have been identified as salient risk factors for adult depression. However, not
Received in revised form 2 December 2016 all individuals who experience ACEs go on to develop depression. The extent to which
Accepted 20 December 2016
resilience- or the ability to demonstrate stable levels of functioning despite adversity- may
Available online 2 January 2017
act as a buffer against depression among individuals with a history of ACEs has not been ade-
quately examined. To address the associations between ACEs, depression, and resilience,
Keywords:
4006 adult participants were recruited from primary care clinics. Participants completed
Adverse childhood experiences
self-report questionnaires including: the Adverse Childhood Experiences Questionnaire, a
Childhood trauma
Consequences of maltreatment retrospective measure of childhood adversity; the Patient Health Questionnaire-9, a mea-
Resilience sure of the presence and severity of the major symptoms of depression; and the Connor
Depression Davidson Resilience Scale, a measure of psychological resilience. Results from regres-
Primary care sion analyses indicated that, while controlling for a range of demographic variables, both
ACEs and resilience independently predicted symptoms of depression, F(9, 3040) = 184.81,
R2 = 0.354. Further, resilience moderated the association between ACEs and depression,
F(10, 3039) = 174.36, p < 0.001, R2 = 0.365. Specifically, the association between ACEs and
depression was stronger among individuals with low resilience relative to those with high
resilience. This research provides important information regarding the relationships among
ACEs, resilience, and depression. Results have the potential to inform the development of
treatments aimed to reduce symptoms of depression among primary care patients with a
history of childhood adversity.
© 2016 Elsevier Ltd. All rights reserved.

1. Introduction

The literature demonstrates robust associations between adverse childhood experiences (ACEs) and ensuing psychosocial
problems (Anda et al., 2002), health-risk behaviours (Dube et al., 2006; Ford et al., 2011), disease (Anda et al., 2008; Dong
et al., 2004), and other undesirable long-term outcomes. One area of particular interest has been the influence of ACEs on the
development and maintenance of depressive symptoms in adulthood. Research in this area suggests that childhood adversity
is predictive of an increased prevalence of lifetime and recent depressive disorders in adulthood (see Heim, Newport, Mletzko,
Miller, & Nemeroff, 2008 for a review).

∗ Corresponding author at: Department of Psychology, University of Calgary, 2500 University Drive NW, Calgary, Alberta T2N 1N4, Canada.
E-mail addresses: jpoole@ucalgary.ca (J.C. Poole), ksdobson@ucalgary.ca (K.S. Dobson), dennis.pusch@albertahealthservices.ca (D. Pusch).

http://dx.doi.org/10.1016/j.chiabu.2016.12.012
0145-2134/© 2016 Elsevier Ltd. All rights reserved.
90 J.C. Poole et al. / Child Abuse & Neglect 64 (2017) 89–100

Although ACEs are widely accepted as an important risk factor for the development of major depressive disorder (MDD),
not all individuals who experience ACEs go on to develop MDD as adults. Individuals who demonstrate stable and healthy
levels of functioning despite experiences of adversity are commonly referred to as “resilient”. Psychological resilience is con-
ceptualized as a multidimensional construct that includes characteristics of tenacity, self-efficacy, emotional and cognitive
control under pressure, adaptability, tolerance of negative affect, and goal orientation (Connor & Davidson, 2003). To date,
no research has adequately explored the role of resilience as a protective factor of depression among adults with a history
of ACEs.

1.1. Major depressive disorder (MDD) in adulthood

MDD is one of the most commonly diagnosed disorders among adults (Kessler et al., 2003) and ranks third among all
disorders responsible for the global disease burden (Mathers & Loncar, 2006). Individuals with MDD experience a range of
symptoms, such as depressed or irritable mood and decreased interest in most activities, and rates of suicide attempts and
completions are elevated among individuals with MDD relative to the population at large (American Psychiatric Association,
2013; Coryell & Young, 2005). Individuals who suffer from MDD also report increased use of social and medical services,
functional impairment (e.g., social and/or work impairment), and lost productivity as a result of impairment at work or
absenteeism (Kessler et al., 2003; Simon, 2003; Wang, Simon, & Kessler, 2006).
Given the tremendous consequences of MDD, it is critical to develop empirically sound models of depression and associ-
ated risk and protective factors. Beck’s (2008) cognitive model of depression proposes that experiences of adversity early in
life represent a risk factor for the formation of dysfunctional attitudes and negative cognitions. When such attitudes and cog-
nitions are activated by daily life events, they produce attentional biases, negatively biased interpretations, and symptoms
of depression. With repeated activation prior to and following depressive episodes, these attitudes and cognitions become
more salient and resistant to change, resulting in chronic and recurring major depressive disorder. Indeed, many contem-
porary theoretical models highlight the role of childhood experiences in the development and maintenance of subsequent
depression (Morris, Kouros, Fox, Rao, & Garber, 2014).

1.2. The association between adverse childhood experiences and depression

Adverse childhood experiences (ACEs) are defined as exposure to childhood emotional, physical, and sexual abuse;
emotional and physical neglect; and household dysfunction (i.e., household substance abuse, mental illness, and crimi-
nal behavior; interparental violence; parental separation or divorce) prior to the age of 18 years. Approximately two thirds
of Americans report exposure to at least one type of childhood adversity and 12% report at least four types of exposures
(Dube et al., 2001). Moreover, the majority of respondents exposed to one type of childhood adversity are typically exposed
to another (Edwards, Holden, Felitti, & Anda, 2003).
Past research has shown strong dose-response relationships between ACEs and symptoms of depression across the
lifespan (Afifi et al., 2008; Chapman et al., 2004; Kendler et al., 2000; Nelson et al., 2002). For instance, Chapman et al. (2004)
found that women who reported five or more types of ACEs were 4.4 times more likely to report recent depressive symptoms
in adulthood than women who reported no ACEs. Dube et al. (2001) found that each ACE increased risk of attempted suicide
across the lifespan 2- to 5-fold. These findings highlight the importance of research that evaluates the cumulative impact of
multiple types of ACEs.

1.3. ACEs, depression, and psychological resilience

A burgeoning field of empirical and theoretical research on protective factors associated with depression may explain
why certain individuals achieve positive developmental outcomes despite exposure to childhood adversity. One such factor
is psychological resilience. Broadly, resilience is conceptualized as the adaptive ability to cope with adversity or trauma
(Bonanno, 2004), and is generally viewed as a multidimensional construct that arises from the interaction among constitu-
tional, biological, cognitive factors, and interpersonal factors (e.g., Caspi et al., 2002; Feder, Nestler, & Charney, 2009; Luthar
& Cicchetti, 2000; Wright, Masten, & Narayan, 2013). Resilience is a dynamic construct and, although it is often seen to pos-
sess trait-like characteristics, longitudinal research suggests that it is modifiable (Montpetit, Bergeman, Deboeck, Tiberio, &
Boker, 2010). Characteristics associated with psychological resilience include the use of active and adaptive coping strategies
(Southwick, Vythilingam, & Charney, 2005), dispositional optimism and positive emotions (Charney, 2004; Ong, Bergeman,
Bisconti, & Wallace, 2006), and purpose in life (Alim et al., 2008; Southwick et al., 2005). The study of resilience holds
the potential to inform practice, prevention, and policy efforts that aim to foster positive adaptation, particularly among
vulnerable populations (Wright et al., 2013).
Only three studies to date have simultaneously examined the associations among childhood adversity, psychiatric symp-
toms (e.g., depression), and resilience (Campbell-Sills, Cohan, & Stein, 2006; Seok et al., 2012; Wingo et al., 2010). While
each study offers unique and valuable findings, all suffer from significant methodological limitations, such as the failure to
capture a wide range of ACEs (i.e., examining only childhood neglect; Campbell-Sills et al., 2006), the utilization of small
sample sizes (Seok et al., 2012), and analysis of samples that are not representative of general populations (Wingo et al.,
 J.C. Poole et al. / Child Abuse & Neglect 64 (2017) 89–100 91

2010). Thus, the existing research leaves several important questions unanswered regarding the association between ACEs,
psychiatric symptoms, and resilience.

1.4. Rationale and significance

A primary goal of the current study was to elucidate risk and resilience factors involved in the development of MDD
among adult primary care patients. Findings from this study are particularly informative for at least two reasons. First, this
research further clarified the association between a prevalent risk factor (i.e., ACEs) and a consequential health outcome
(i.e., depression). Second, this research examined a potential buffer of the impact of ACEs on depression. Given that ACEs
represent a distal- or historical- risk factor in the development of depression, it is critical that researchers identify potential
targets for treatment. Indeed, empirical findings that improve our understanding of the factors and processes involved in
positive adaptation despite adversity are invaluable at the personal, familial, and societal levels, particularly in light of
current concerns regarding rising costs of global healthcare.

1.5. Research hypotheses

The current study evaluated two primary hypotheses. First, it was hypothesized that there would be a positive association
between cumulative ACEs and symptoms of depression. Second, it was hypothesized that the association between ACEs and
depression would vary as a function of resilience, such that the association between ACEs and symptoms of depression
would be stronger among individuals with low resilience than among individuals with high resilience.

2. Methods

2.1. Participants

The current study was part of the EmbrACE Study, a large-scale research program that investigated the relationship
between ACEs and physical and mental health outcomes among adult primary care patients in Calgary, Alberta, Canada.
The EmbrACE Study was approved by the University of Calgary’s Human Research Ethics Board and collected data from a
convenience sample of 4006 primary care patients aged 18 years and older. Participants were recruited from 11 primary
care clinics in the greater Calgary area. Recruitment took place from October 2014 to July 2015.

2.2. Procedure

Primary care physicians at clinics in Calgary, AB and surrounding areas were invited to participate in the EmbrACE Study.
Primary care settings that participated in the study allowed EmbrACE Study team members to be in their clinic. Each clinic
was paid a fee ($1000 CAD) in consideration of the staff time and disruption that the study entailed. Recruitment at each
clinic lasted, on average, approximately six weeks.
During recruitment, trained research assistants approached patients in clinic waiting areas after patients had registered
at the front desk and were waiting to speak to their physician concerning a myriad of health-related issues, including general
check-ups and various mental/physical health concerns. Following a brief description of the study, written informed consent
and contact information was obtained from interested patients. Patients were offered a choice between an online or paper
format to complete the study questionnaire package. Patients who specified a preference for the online survey (n = 2737)
were provided with a flash card that outlined instructions to access the online survey, which expired two weeks following
initiation. Individuals who specified a preference to complete the paper survey (n = 1269) were provided with the survey
and a pre-addressed and pre-paid envelope to mail the completed survey back to the research team. All participants were
provided with a $25.00 gift card following completion of the questionnaire package.

2.3. Measures

2.3.1. Demographic information. Information on age, gender, ethnicity, education, annual household income, marital status,
and employment status was obtained from all participants. The variables of ethnicity, education, income, marital status, and
employment status were dummy coded.

2.3.2. Adverse childhood experiences (ACEs). Experiences of childhood adversity were measured using the Adverse Childhood
Experiences (ACE) Questionnaire, a 29-item scale adapted from a variety of published questionnaires including the Conflict
Tactics Scale (Straus, 1979), The Child Trauma Questionnaire (Bernstein et al., 2003), and the Wyatt (1985) questions on
sexual abuse. Response options to items include “Never”, “Once or twice”, “Sometimes”, “Often, or ‘Very Often’. The 10 types
of ACEs (i.e., emotional, physical, and sexual abuse; emotional and physical neglect; five types of household dysfunction)
were each coded as a binary variable (occurred: “1”, did not occur: “0”). The total number of ACEs reported by participants
represents the “total ACE score” (range: 0–10), which was used to assess the cumulative effect of multiple ACEs.
92 J.C. Poole et al. / Child Abuse & Neglect 64 (2017) 89–100

The psychometric properties of the ACE Questionnaire were assessed during an earlier phase of the EmbrACE study and
found to be appropriate for use with a primary care sample, as it demonstrated excellent internal consistency (Cronbach’s
␣ = 0.95) and construct validity (correlated with both the Child Abuse and Trauma Scale (CATS; r = 0.94), and the Childhood
Trauma Questionnaire (CTQ; r = 0.95) ([reference removed for blind review])). Previously reported measures of test-retest
reliability have reported substantial reliability of the ACE Questionnaire (Kappa = 0.64) (Dube, Williamson, Thompson, Felitti,
& Anda, 2004; Sim & Wright, 2005).

2.3.3. Symptoms of depression. Symptoms of depression were measured with the Patient Health Questionnaire-9 (PHQ-9),
a self-rated 10-item scale used to assess the presence and severity of the nine major symptoms of depression as outlined
by the Diagnostic and Statistical Manual of Mental Disorders, 5th edition (DSM-5) (American Psychiatric Association, 2013;
Spitzer, Kroenke, & Williams, 1999). Items ask respondents to rate their experience of each symptom of depression over the
last 2 weeks according to the four response options of “not at all”, “several days”, “more than half the days”, and “nearly
every day”. The final question asks respondents “how difficult have these problems made it for you to do your work, take
care of things at home, or get along with other people?” and is used to assess impaired functioning as a result of depression.
Total scores on the PHQ-9 are based on the first 9 items and range from 0 to 27. Higher scores indicate greater severity of
depression.
The PHQ-9 can be utilized as a binary indicator of MDD according to criteria outlined by the DSM-5 (American Psychiatric
Association, 2013). To meet the diagnostic criteria for MDD, respondents must endorse the following: 1) either “little interest
or pleasure in doing things” or “feeling down, depressed or hopeless” for more than half the days or nearly every day, 2)
a total of five symptoms for more than half the days or nearly every day, and 3) an indication that these problems have
caused significant impairments in work, home, or social functioning. Use of both the continuous total PHQ-9 score and the
binary score have been validated among primary care samples and demonstrated excellent psychometric properties (e.g.,
Cronbach’s ␣ = 0.89; test-retest reliability r = 0.84, 48-h; construct validity as indicated by associations with other validated
measures of depression) (Kroenke, Spitzer, & Williams, 2001; Löwe et al., 2004; Williams, Noel, Cordes, Ramirez, & Pignone,
2002).

2.3.4. Psychological resilience. Psychological resilience was measured with the 10-item Connor-Davidson Resilience Scale
(CDRISC), a self-report measure that assesses core aspects of psychological resilience such as tolerance of negative affect,
ability to focus under pressure, and an action-oriented approach to problem solving (Connor & Davidson, 2003). Respondents
rate the degree to which characteristics in response to challenge apply to them (e.g., “able to adapt to change”), from “not
at all” (0) to “true nearly all the time” (4). Total scores range from 0 to 40, where higher scores indicate greater resilience.
Previous studies have reported that the CDRISC demonstrates excellent internal consistency (Cronbach’s ␣ = 0.85), test-retest
reliability (r = 0.87), and discriminant and convergent validity (Campbell-Sills et al., 2006; Campbell-Sills & Stein, 2007). The
10-item CDRISC correlates highly with the original 25-item CDRISC (r = 0.92; Campbell-Sills & Stein, 2007).

3. Statistical analyses

3.1. Data screening and management

All variables were screened for missing values and outliers and possible statistical assumption violations. Participants
who endorsed the gender option of “other” were removed from analyses due to their small representation in the sample
(n = 8). An analysis of standard residuals identified a total of 34 outliers on the main variables of interest (i.e., ACE, CDRISC,
and PHQ-9 scores) and these cases were removed from the data set. Additional tests indicated that the data met assumptions
related to multicollinearity, linearity, homoscedasticity, and normally distributed errors. List-wise deletion was utilized to
handle all missing data. Finally, one-way ANOVAs were utilized to evaluate differences in responses to self-reported data as
a function of survey modality (i.e., online vs. paper questionnaire). Results indicated that there were statistically significant
differences (p < 0.05) between those who completed the survey in the online format (n = 2737) and paper format (n = 1269)
on each main variable of interest (i.e., ACEs, CDRISC, PHQ-9). However, with large samples such as that of the current study,
even negligible differences in outcome variables can demonstrate statistical significance. As such, estimates of effect sizes
are utilized as a more accurate measure of significance among large samples (Sullivan & Feinn, 2012). In the current study, all
eta squared effect sizes for differences on the ACEs (␩2 = 0.001), CDRISC (␩2 = 0.013), and PHQ-9 scales (␩2 = 0.003) were of
small practical significance (Cohen, 1988). As such, analyses were conducted on the sample as a whole, regardless of survey
mode.

3.2. Main analyses

All analyses were performed using IBM SPSS Software, version 21. Descriptive variables were: 1) calculated for the sample
as a whole, and 2) compared between those who reported depressive symptoms that met the diagnostic criteria for MDD
and those who did not. To evaluate the internal consistency of the PHQ-9 and CDRISC measures with the [removed for blind
review] Study sample, Cronbach’s ␣ (Cronbach, 1951) was calculated for each measure.
 J.C. Poole et al. / Child Abuse & Neglect 64 (2017) 89–100 93

The hypothesis that there would be a positive association between ACEs and depression was examined in two different
ways: first with depression as a continuous variable, and second with depression as a binary variable. In the first set of anal-
yses, continuous scores on the PHQ-9 were entered as the dependent measure in a two-stage hierarchical linear regression.
In the first block of the model, age, gender, ethnicity, education, household income, marital status, and employment were
simultaneously entered as predictors to control for possible effects of covariates. In the second block of the model, total
ACE score was entered as the primary predictor of interest. In the second set of analyses, adjusted odds ratios (ORs) and
95% confidence intervals (CIs) were calculated from separate logistic regression models to evaluate the increased likelihood
of reporting MDD with increases in total ACE score (i.e., 1–4 or more total ACEs), and the increased likelihood of reporting
MDD for each of the ten types of ACEs. In all logistic regression models, zero ACEs acted as the referent category. All logis-
tic regression models included age, gender, ethnicity, education, income, marital status, and employment as predictors to
control for covariates.
To evaluate the hypothesis that resilience would moderate the association between ACEs and symptoms of depression,
hierarchical regression analyses were utilized to evaluate the interaction effect between total ACEs and total resilience on
total symptoms of depression. In preparation for the analysis and in accord with the recommendations of Aiken and West
(1991), the two predictors (ACE and resilience score) were centered prior to analyses. Analyses included age, ethnicity,
education, income, marital status, and employment as predictors to control for covariates.

4. Results

Table 1 summarizes the means and standard deviations for all measured variables and descriptive data for the sample.
Approximately 68% of the respondents were women. Participants ranged in age from 18 to 92 years (M = 44 years, SD = 16.98).
Eighty three percent of respondents were White, 1% were Black, 7.3% were Asian, 0.9% Native American, and 5.3% were of other
ethnicities. In general, the sample was of relatively high socioeconomic status and was highly educated (e.g., approximately
57% of the sample had obtained post-secondary or graduate degrees). Most respondents (64.2%) were married or living with
their partner. Table 1 includes separate demographic variables for participants who met the criteria for MDD (7.4%) and for
those who did not (92.6%). As can be seen, the two groups differed significantly on all variables with the exception of gender,
␹2 (1) = 2.28, p =0.13.
Pearson’s correlations were calculated for the main variables of interest. Depression was significantly associated with both
the total number of ACEs, r = 0.35, p < 0.001, and resilience, r = −0.50, p < 0.001. The total number of ACEs and resilience were
also significantly correlated, r = −0.19, p < 0.001, however, a correlation coefficient of this size indicates little or no relationship
among large samples (Colton, 1974). Using the [removed for blind review] Study sample, Cronbach’s ␣ (Cronbach, 1951)
was calculated for the PHQ-9 (9 items), ␣ = 0.89, and the CDRISC (10 items), ␣ = 0.93. These values indicate excellent internal
consistency of these measures (George & Mallery, 2003).
To evaluate the hypothesis that total ACE score would be positively associated with symptoms of depression, hierarchical
regression analyses were conducted with symptoms of depression as a continuous outcome. When the covariates were
entered on the first block, the prediction model was statistically significant, F(7,3097) = 58.79, p < 0.001, R2 = 0.117, Adjusted
R2 = 0.115. Gender did not predict depression, however age, ethnicity, education, income, marital status, and employment
did (p < 0.05). When ACE score was entered on the second block, the model increased substantially in its predictive power,
F(8,3096) = 105.02, p < 0.001, R2 = 0.213, Adjusted R2 = 0.211. Thus, as indexed by the R2 statistic, the addition of ACE score
to the model accounted for 9.6% of explained variance in depressive symptoms, beyond the effects of the covariates. Of all
predictors in the model, ACE score was shown to have the strongest relationship to depressive symptoms (␤ = 0.32). Thus,
as hypothesized, a greater number of total ACEs predicted a greater number of symptoms of depression, even with all other
predictors in the analysis statistically controlled (see Fig. 1).
Separate logistic regression analyses were then used to examine how ACEs increase the likelihood of meeting diagnostic
criteria for MDD. Results indicated that each of the predictor models provided a statistically significant prediction of MDD,
as chi-square values ranged from ␹2 (8, N = 3241) = 161.00, p < 0.001 to ␹2 (8, N = 3241) = 209.95, p < 0.001. The Nagelkerke
pseudo R2 values indicated that the models accounted for approximately 11%–25% of the total variance in MDD diagnosis.
Table 2 presents the prevalence rates, Wald test values, odds ratios [Exp(B)], and the 95% confidence intervals (CI) for the
odds ratios for total ACE score and for each type of ACE. The Wald test indicated that individuals with a history of one type of
ACE were no more likely to report MDD than individuals with no history of ACEs, p > 0.05. Individuals with two types of ACEs,
however, were 3.13 times (CI = 1.82–5.39), p < 0.001, more likely to report MDD than those with no ACEs, and individuals
with three types of ACEs were 4.54 times (CI = 2.59–7.97), p < 0.001, more likely to report MDD than those with no ACEs.
Finally, those with four or more types of ACEs were 7.25 times (CI = 4.48–11.72), p < 0.001, more likely to report MDD than
those with no ACEs.
Analyses also indicated that each type of ACE was a statistically significant predictor of MDD (p < 0.001). As compared to
the association between other types of ACEs and MDD, the association between emotional abuse and MDD appeared to be
particularly salient. Specifically, individuals with a history of emotional abuse were 3.51 times (CI = 2.62–4.69) more likely
than those without a history of emotional abuse to meet the criteria for MDD, adjusting for age, gender, ethnicity, education,
income, marital status, and employment status.
To evaluate the hypothesis that resilience would moderate the association between ACEs and symptoms of depression, a
hierarchical linear regression analysis was used (Table 3). On the first step, the covariates of age, gender, ethnicity, education,
94 J.C. Poole et al. / Child Abuse & Neglect 64 (2017) 89–100

Table 1
Demographic characteristics of the sample.

Entire Sample Non-Depressed Depressed Significance Test

(Non-Depressed vs.
Depressed)

Variable n %/M(SD) n %/M(SD) n %/M(SD) ␹2 /t p

Age
Mean 3509 44.13(16.98) 3139 44.45(16.93) 251 37.80(15.23) 6.03 <0.001***

Gender
Male 1269 31.8% 1131 32.0% 90 28.0% 2.28 0.13
Female 2719 68.2% 2396 67.9% 231 72.0%

Education
Less than high school 201 5.0% 164 4.7% 29 9.0% 28.08 <0.001***
High school or equiv. 612 15.4% 525 14.9% 58 18.1%
Some post-secondary 892 22.4% 776 22.0% 87 27.2%
Post-secondary 1872 47.0% 1685 47.8% 123 38.5%
Graduate degree 408 10.2% 376 10.6% 23 7.2%

Income
Less than 20,000 395 10.1% 289 8.4% 86 27.3% 142.40 <0.001***
20,000–39,999 480 12.3% 414 11.9% 45 14.3%
40,000–59,999 573 14.6% 493 14.2% 57 18.1%
60,000–79,999 542 13.8% 489 14.1% 40 12.7%
Greater than 80,000 1924 49.2% 1781 51.4% 87 27.6%

Marital Status
Married/Common Law 2557 64.2% 2342 66.5% 130 40.5% 136.77 <0.001***
Never married 911 22.9% 752 21.4% 131 40.8%
Widowed 107 2.7% 95 2.7% 5 1.6%
Separated/Divorced 405 10.2% 331 9.4% 55 17.2%

Employment
Full-Time 1859 46.7% 1670 47.5% 131 40.8% 53.66 <0.001***
Part-Time 848 21.3% 724 20.6% 91 28.3%
Unemployed 617 15.5% 519 14.8% 80 24.9%
Retired 654 16.4% 605 17.2% 19 5.9%

Ethnicity
Caucasian 3302 83.0% 2931 83.2% 262 81.9% 30.15 <0.001***
African American 39 1.0% 36 1.1% 2 0.6%
Asian 289 7.3% 344 9.7% 32 10.0%
First Nations 35 0.9% 22 0.7% 11 3.4%
Other 209 5.3% 188 5.3% 13 4.1%

ACES
Total score 3686 1.92(2.12) 3315 1.79(2.02) 267 3.55(2.71) −13.30 <0.001***

PHQ-9
Total score 3821 4.84 (5.08) 3532 3.82(3.64) 289 17.24(3.51) −62.28 <0.001***

CDRISC
Total score 3836 30.27 (7.07) 3413 30.97 277 22.29 (7.65) 27.46 <0.001***

Note: ***p < 0.001; ACEs = total ACEs (0–10). CDRISC = Connor Davidson Resilience Scale.

income, marital status, and employment status were entered and significantly predicted depression, F(7, 3042) = 48.35,
p < 0.001, R2 = 0.100, adjusted R2 = 0.098. When the ACEs and resilience variables were entered on the second block, the model
explained an additional 25.4% of variance in symptoms of depression while controlling for the covariates, F(9, 3040) = 184.81,
R2 = 0.354, adjusted R2 = 0.352. When the ACEs X resilience interaction term was entered on the third block, the model
explained an additional 1.1% of variance in symptoms of depression, F(10, 3039) = 174.36, p < 0.001, R2 = 0.365, adjusted
R2 = 0.362 (Table 3).
Given that the ACEs X resilience interaction significantly predicted symptoms of depression, two additional analyses were
performed treating resilience as the moderator variable. The first analysis was carried out at a resilience level of +1 SD unit
by re-centering the resilience variable at that value; the intercept and raw regression coefficient for ACEs were 5.25 and 0.34
(SE = 0.05), respectively. The second analysis was carried out at a resilience level of −1 SD unit by centering the resilience
variable at that value; the intercept and raw regression coefficient for ACEs were 9.18 and 0.76 (SE = 0.042), respectively. As
can be seen in Fig. 2, symptoms of depression were higher with increasing number of ACEs across the range of resilience,
but the rate of this increase was more pronounced at lower levels of resilience. Generally, higher resilience acts as a buffer
to lower depression with increased rates of ACEs.
 J.C. Poole et al. / Child Abuse & Neglect 64 (2017) 89–100 95

Fig. 1. The relationship between total ACE score and mean PHQ-9 score.

Table 2
Prevalence and adjusted odds ratio of adverse childhood experiences (ACEs) endorsed by participants without depression and with depression.

Total ACEs Reported No MDD, % (n) MDD, % (n) Wald AOR 95% CI

0 ACEs 32.2% (1067) 11.6% (31) – – –

1 ACE 24.6% (814) 14.2% (38) 2.12*** 1.53 .87–2.71
2 ACEs 16.8% (557) 16.1%(43) 16.99*** 3.13 1.82–5.39
3 ACEs 10.0% (332) 14.2% (38) 27.87*** 4.54 2.59–7.97
4 or more ACEs 16.4% (545) 43.8% (117) 65.27*** 7.25 4.48–11.72

Type of ACE Reported

Emotional Abuse 14.2% (487) 39.4% (111) 71.53*** 3.50 2.62–4.67
Emotional Neglect 10.4% (359) 28.8% (80) 45.42*** 3.00 2.18–4.13
Physical Abuse 12.1% (418) 27.4% (77) 37.11*** 2.73 1.97–3.76
Physical Neglect 6.5% (225) 16.5% (46) 21.13*** 2.51 1.69–3.71
Sexual Abuse 19.5% (675) 36.7% (104) 46.60*** 2.81 2.09–3.79
Interparental Violence 12.0% (415) 27.0% (76) 25.50*** 2.29 1.66–3.16
Household Substance Abuse 30.0% (1037) 47.0% (133) 32.46*** 2.24 1.70–2.96
Parental Separation/Divorce 25.2% (873) 43.3% (122) 21.78*** 1.95 1.47–2.57
Household Incarceration 7.9% (274) 19.3% (54) 21.33*** 2.37 1.64–3.42
Household Mental Illness 43.3% (1498) 71.6% (202) 54.87*** 3.20 2.35–4.35

Note: df = 1. The dependent variable was MDD with MDD as the target category and no MDD as the reference category. Each logistic regression analysis
adjusted for age, gender, ethnicity, education, income, marital status, and employment status. AOR indicates adjusted odds ratio; 95% CI (confidence
interval) shown in parentheses; The Nagelkerke R2 values ranged from 0.11 to 0.25.

5. Discussion

The current study evaluated retrospective reports of adverse childhood experiences (ACEs) as a risk factor for recent
depression in a large sample of adult primary care patients. Results indicated that the majority of participants (70%) were
exposed to at least one type of ACE, and that such exposure increased the risk for depression. Consistent with study hypothe-
ses and previous research, participants who reported greater cumulative exposure to ACEs were more likely to report
increased symptoms of depression. This study also expanded on existing research by evaluating the role of psychological
resilience as a buffer against depression for adults with a history of ACEs. Resilience moderated the relationship between ACEs
and depression, as cumulative ACEs were more strongly associated with total symptoms of depression among individuals
with low resilience than among those with high resilience.
Approximately 7.6% of respondents in the current study reported recent symptoms of depression that met the diagnostic
criteria for MDD. Although prevalence rates of MDD among
Canadian primary care patients have not been reported, these rates were consistent with those reported among primary
care samples in the United States and New Zealand (range from 6.6% to 8.6%; Arroll et al., 2010; Kessler et al., 2003). These
prevalence rates reinforce the importance of the identification, assessment, and treatment of depression within primary
care settings.
96 J.C. Poole et al. / Child Abuse & Neglect 64 (2017) 89–100

Table 3
Hierarchical regression model for Patient Health Questionnaire-9 (PHQ-9).

Block R2 Model b SE-b Beta sr2

1 0.100 (Constant)*** 8.707 0.855

Age*** −0.068 0.006 −0.227 0.040
Gender 0.297 0.188 0.028 0.001
Ethnicity −0.435 0.239 −0.032 0.001
Education** −0.762 0.222 −0.060 0.003
Income*** −1.062 0.188 −0.106 0.009
Marital status*** 1.251 0.198 0.120 0.012
Employment 0.269 0.202 0.025 0.001

2 0.354 (Constant)*** 6.489 0.728

Age*** −0.037 0.005 −0.123 0.011
Gender −0.280 0.160 −0.026 0.001
Ethnicity −0.280 0.203 −0.021 0.000
Education −0.073 0.190 0.006 0.000
Income** −0.544 0.160 −0.054 0.003
Marital status*** 1.112 0.168 0.107 0.009
Employment 0.131 0.172 0.012 0.000
Total ACEs*** 0.596 0.035 0.256 0.062
Total Resilience*** −0.296 0.011 −0.411 0.152

3 0.365 (Constant)*** 6.247 0.723

Age*** −0.037 0.005 −0.124 0.011
Gender −0.254 0.159 −0.024 0.001
Ethnicity −0.257 0.201 −0.019 0.000
Education −0.045 0.188 −0.004 0.000
Income** −0.494 0.159 −0.049 0.002
Marital status*** 1.113 0.167 0.107 0.009
Employment 0.110 0.171 0.010 0.000
Total ACEs*** 0.556 0.035 0.239 0.052
Total Resilience*** −0.286 0.011 −0.396 0.140
ACE X Resilience*** −0.031 0.004 −0.108 0.011

Note: sr2 is the squared semi-partial correlation.

**
p < 0.01.
***
p < 0.001.

Fig. 2. The effect of adverse childhood experiences on depressive symptoms in adulthood and the moderating effect of resilience.

Respondents in this study who met the diagnostic criteria for MDD reported higher total ACE scores than those who
did not meet the criteria for MDD, and the positive association between ACE score and depressive symptoms remained
significant even when other personal and social factors were held constant. These findings are consistent with research
that has repeatedly identified associations between early life trauma or adversity and subsequent depression (Heim et al.,
2008). Differences in the prevalence rates of ACEs reported by study participants who met the criteria for MDD as compared
to those who did not were particularly striking. For instance, while 32.2% of participants without depression reported no
exposure to childhood adversity, only 12.7% of participants with depression reported the same. Conversely, while only 16.4%
 J.C. Poole et al. / Child Abuse & Neglect 64 (2017) 89–100 97

of participants without depression reported exposure to four or more ACEs, 41% of participants with depression reported the
same. Indeed, respondents’ likelihood of reporting MDD increased dramatically as their ACE score increased. As compared
to individuals who reported no history of ACEs, those who reported two ACEs were more than 3 times as likely to report
MDD, those who reported three ACEs were more than 4.5 times as likely to report MDD, and those who reported four or
more ACEs were over 7 times as likely to report MDD.
Although all types of ACEs were predictive of MDD, the association between emotional abuse and MDD was greater than
that of any other type of ACE and MDD. Participants who were exposed to emotional abuse as children were almost 3.5
times more likely to meet the diagnostic criteria for MDD than those who were not exposed to emotional abuse. This result
is consistent with that reported in previous research, such as rates of depression that were 3.1 and 3.3 times higher among
women and men who reported emotional abuse, respectively, than those who did not (Chapman et al., 2004). The notion that
childhood emotional abuse is a more potent risk factor for depression than other types of childhood adversity is consistent
with cognitive theoretical frameworks. For instance, Rose and Abramson (1992) posit that repeated exposure to emotional
abuse provides direct depressive cognitions from abusers (e.g., “you’re worthless”) that contribute to the development
of cognitive attributions most consistent with depression (e.g., “I am worthless”). Alternatively, abusers who perpetrate
physical or sexual abuse may not provide direct attributions regarding the cause of the abuse, and therefore victims may be
less likely to make negative attributions about themselves (Rose & Abramson, 1992).
A novel contribution of the current study was the exploration of psychological resilience as a protective factor against
depression among individuals with a history of childhood adversity. The current study was the first to report rates of resilience
as captured by the CDRISC in a Canadian primary care sample. Results indicate that levels of resilience were comparable
(range: 0–40; median = 31) to those reported by American primary care samples (range: 0–40; median = 35) (Wingo et al.,
2010). As hypothesized, psychological resilience was associated with lower symptoms of depression among adults with a
history of childhood adversity. Although total ACE scores were positively associated with symptoms of depression across
the range of participants, this association was stronger for individuals with low resilience than for those with high resilience.
This result was inconsistent with those found by Wingo et al. (2010), who reported that psychological resilience did
not moderate the association between childhood adversity and depression. However, this discrepancy must be considered
in light of several key differences between the former and current study. For instance, participants in Wingo et al. (2010)
study were almost entirely African American (94%) and low-income (63% reported a monthly income of <$1000) and thus
differed from the predominantly Caucasian, high-income sample of the current study. Such differences may be important, as
resilience appears to differ as a function of both ethnicity (Campbell-Sills et al., 2006) and income (Wagnild, 2003). Further,
the participants in Wingo and colleague’s study were recruited from both primary care clinics and obstetrical-gynecological
clinics, but the proportion of the sample recruited from each type of clinic was not stated, nor was the proportion of pregnant
women in the sample. It is likely that individuals who face major life changes such as pregnancy, may respond differently
to questionnaires about childhood trauma, resilience, and depression than individuals who are not undergoing such life
changes. Finally, Wingo et al. (2010) reported a median of zero experiences of childhood abuse in their sample. Low rates
of child abuse among this sample may reflect the authors’ utilization of a self-report measure that did not capture a wide
range of ACEs (i.e., did not measure childhood neglect or household dysfunction), or the possibility that child abuse may be
normalized and therefore underreported among groups of ethnic minorities in which ACEs are more common (Finkelhor,
Ormond, Turner, & Hamby, 2005).
While the present study found that more resilient individuals reported lower levels of depressive symptoms when ACEs
were also low, a previous study found that highly resilient individuals who experienced high levels of emotional neglect
reported the lowest levels of current psychological distress, even compared to highly resilient individuals who experienced
low levels of emotional neglect (Campbell-Sills et al., 2006). This discrepancy may be due to the fact that the earlier study
examined only one type of ACE (i.e., neglect), whereas the current study evaluated the cumulative effect of multiple types
of ACEs. Alternatively, the interactive effect of resilience and childhood adversity may differ for depression as compared
to general psychological distress. Future research should evaluate the role of various ACEs and their impact on diverse
psychological disorders as a function of resilience and childhood adversity.

5.1. Strengths and limitations

The current study was based on strong theoretical rationale and thus the results have important theoretical implications
(e.g., how ACEs, resilience, and depression are conceptualized within primary care settings) and practical implications (e.g.,
how ACEs, resilience, and depression are assessed and treated within primary care settings). Additional strengths of this
research include analyses of a large sample highly relevant to health-care providers, the evaluation of a wide range of ACEs,
and the selection of assessment measures with strong psychometric properties.
Despite these strengths, the results should be considered in light of several limitations. First, the reliance on a cross-
sectional design precludes inferences about causality. For instance, it is possible that the onset of depressive symptoms
may have occurred prior to exposure to ACEs, in which case it would be inaccurate to conclude that ACEs are predictive of
depression. However, the strong associations between ACEs and recent symptoms of depression do mitigate this concern.
Relatedly, while it was hypothesized that increased resilience would be associated with decreased symptoms of depression
among individuals with a history of childhood adversity, it is possible that symptoms of depression cause individuals to
98 J.C. Poole et al. / Child Abuse & Neglect 64 (2017) 89–100

perceive themselves as less resilient. Future research with prospective or longitudinal designs would clarify the direction
and temporal order of relationships among ACEs, resilience, and depression.
Second, the reliance on self-report data also represents a limitation of the current research, as self-report measures are
susceptible to recall bias or deficits. However, each of the measures utilized has demonstrated strong psychometric prop-
erties. Further, longitudinal follow-up of adults who reported childhood abuse has demonstrated that reports of childhood
abuse often underestimate actual occurrence, which might attenuate the association between ACEs and depression (Hardt
& Rutter, 2004). Relatedly, while it is possible that current mood may bias the recall of childhood events or that respondents
may have difficulty recalling childhood events, previous research has reported good test-retest reliability of the ACE measure
(kappa = 0.64; Dube et al., 2004). Ultimately, it can argued that an individual’s perception of his or her ACEs, resilience, and
depressive symptoms may be of greater consequence in the assessment and treatment of adult health complaints than the
objective reliability of such reports.
Third, results of the current study may have been influenced by unmeasured factors that were associated with the main
variables of interest (i.e. confounding variables). For instance, the current study did not include assessments of current,
comorbid health concerns and health risk behaviours, which have both been shown to influence the onset and course of
depression (Chapman, Perry, & Strine, 2005; Green & Pope, 2000). Protective factors during childhood, such as social support
from caregiver and peers, have also been identified as salient predictors of subsequent mental health outcomes among
individuals who report childhood adversity (Collishaw et al., 2007; Lynskey & Fergusson, 1997), but these factors were not
considered in the current study.
Finally, while the current study focused on psychological resilience, broader definitions of resilience encompass other
influences, such as biological and genetic factors (e.g., polymorphism in monoamine oxidase A genotype, stress-reactivity),
personality factors (e.g., self-esteem, intelligence), and interpersonal factors (social support) (Caspi et al., 2002; Rutter, 2006;
Werner & Smith, 2001). Future research that evaluates which aspects of resilience seem to be most salient in modifying
the relationship between ACEs and depression would offer further insights to how symptoms of depression may be most
effectively prevented and treated among individuals with a history of ACEs.

5.2. Implications and future directions

The accumulated evidence that childhood adversity is a risk factor for depression, particularly among utilizers of primary
care services, supports the application of a developmental model in the assessment and treatment of adult mental health
complaints. For instance, health care providers must be aware of the relevance of ACEs to the assessment of depression among
adult patients. Further, the finding that psychological resilience appears to be a meaningful predictor of successful adaptation
following ACEs holds immense potential to guide treatment programs. Previous research has demonstrated that many of the
characteristics associated with psychological resilience can be enhanced through practice and training (Steinhardt & Dolbier,
2008). Indeed, resilience-training programs that focus on fostering traits of psychological resilience such as the promotion
of positive emotions, cognitive flexibility, and active problem solving and coping skills have been shown to improve rates
of depressive symptoms (e.g., Brunwasser, Gillham, & Kim, 2009). Further, resilience-training programs that foster personal
characteristics required to effectively deal with unanticipated stressors may be important in the management and prevention
of subsequent symptoms of depression, thus minimizing potential for relapse.
Although programs aimed to improve traits of psychological resilience such as those captured in the current study appear
promising, such programs have not yet been evaluated in primary care settings. As such, future researchers may do well to
develop and evaluate an evidence-based treatment protocol that is suitable for implementation within primary care settings,
recognizes the long-term and cumulative effects of ACEs, and emphasizes the development of psychological resilience. It
will be important to demonstrate that such a program is efficacious, cost-effective, and transferable to a range of settings
and populations. Further, future research should evaluate the mechanisms by which resilience-training programs improve
depression, as they currently remain relatively unclear (Brunwasser et al., 2009). Such research will provide important
insights regarding how successful adaptation may be promoted among individuals with a history of exposure to adversity.

Ethical approval

All procedures performed in studies involving human participants were in accordance with the ethical standards of
the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or
comparable ethical standards. This research did not involve animal participants.

Conflict of interest

Poole declares no conflict of interest.

 J.C. Poole et al. / Child Abuse & Neglect 64 (2017) 89–100 99

Acknowledgments

This research was supported by a grant from the Palix Foundation awarded to Dobson and Pusch [RSO number: 1031803,
2015-2016].

References

Afifi, T. O., Enns, M., Cox, B., Asmundson, G. J., Stein, M. B., & Sareen, J. (2008). Population attributable fractions of psychiatric disorders and suicide
ideation and attempts associated with adverse childhood experiences. American Journal of Public Health, 98(5), 946.
Aiken, L. S., & West, S. G. (1991). Multiple regression: Testing and interpreting interactions. Newbury Park, CA: Sage.
Alim, T., Feder, A., Graves, R., Wang, Y., Weaver, J., Westphal, M., . . . & Charney, D. (2008). Trauma, resilience, and recovery in a high-risk African-American
population. American Journal of Psychiatry, 165(12), 1566–1575.
American Psychiatric Association (APA). (2013). Diagnostic and Statistical Manual of Mental Disorders (5th ed.). Washington, DC: American Psychiatric
Publishing.
Anda, R. F., Chapman, D. P., Felitti, V. J., Edwards, V., Williamson, D. F., Croft, J. B., & Giles, W. H. (2002). Adverse childhood experiences and risk of paternity
in teen pregnancy. Obstetrics and Gynecology, 100(1), 37–45.
Anda, R. F., Brown, D. W., Dube, S. R., Bremner, J. D., Felitti, V. J., & Giles, W. H. (2008). Adverse childhood experiences and chronic obstructive pulmonary
disease in adults. American Journal of Preventive Medicine, 34(5), 396–403.
Arroll, B., Goodyear-Smith, F., Crengle, S., Gunn, J., Kerse, N., Fishman, T., . . . & Hatcher, S. (2010). Validation of PHQ-2 and PHQ-9 to screen for major
depression in the primary care population. The Annals of Family Medicine, 8(4), 348–353.
Beck, A. (2008). The evolution of the cognitive model of depression and its neurobiological correlates. American Journal of Psychiatry, 165(8), 969–977.
Bernstein, D. P., Stein, J. A., Newcomb, M. D., Walker, E., Pogge, D., Ahluvalia, T., . . . & Zule, W. (2003). Development and validation of a brief screening
version of the Childhood Trauma Questionnaire. Child Abuse and Neglect, 27(2), 169–190.
Bonanno, G. A. (2004). Loss, trauma, and human resilience: Have we underestimated the human capacity to thrive after extremely aversive events?
American Psychologist, 59(1), 20.
Brunwasser, S. M., Gillham, J. E., & Kim, E. S. (2009). A meta-analytic review of the Penn Resiliency Program’s effect on depressive symptoms. Journal of
Consulting and Clinical Psychology, 77(6), 1042.
Campbell-Sills, L., & Stein, M. B. (2007). Psychometric analysis and refinement of the Connor-Davidson Resilience Scale (CD-RISC): Validation of a 10-item
measure of resilience. Journal of Traumatic Stress, 20(6), 1019–1028.
Campbell-Sills, L., Cohan, S. L., & Stein, M. B. (2006). Relationship of resilience to personality, coping, and psychiatric symptoms in young adults. Behaviour
Research and Therapy, 44(4), 585–599.
Caspi, A., McClay, J., Moffitt, T. E., Mill, J., Martin, J., Craig, I. W., & Poulton, R. (2002). Role of genotype in the cycle of violence in maltreated children.
Science, 297(5582), 851–854.
Chapman, D. P., Whitfield, C. L., Felitti, V. J., Dube, S. R., Edwards, V. J., & Anda, R. F. (2004). Adverse childhood experiences and the risk of depressive
disorders in adulthood. Journal of Affective Disorders, 82(2), 217–225.
Chapman, D. P., Perry, G. S., & Strine, T. W. (2005). The vital link between chronic disease and depressive disorders. Preventing Chronic Disease, 2(1), A14.
Charney, D. S. (2004). Psychobiological mechanisms of resilience and vulnerability: Implications for successful adaptation to extreme stress. FOCUS: The
Journal of Lifelong Learning in Psychiatry, 2(3), 368–391.
Cohen, J. (1988). Statistical analysis for the behavioral sciences. Hillsdale: Lawrance Erlbaum.
Collishaw, S., Pickles, A., Messer, J., Rutter, M., Shearer, C., & Maughan, B. (2007). Resilience to adult psychopathology following childhood maltreatment:
evidence from a community sample. Child Abuse & Neglect, 31(3), 211–229.
Colton, T. (1974). Regression and correlation. Statistics in Medicine, 189.
Connor, K. M., & Davidson, J. R. (2003). Development of a new resilience scale: The Connor-Davidson resilience scale (CD-RISC). Depression and Anxiety,
18(2), 76–82.
Coryell, W., & Young, E. A. (2005). Clinical predictors of suicide in primary major depressive disorder. Journal of Clinical Psychiatry, 66(4), 412–417.
Cronbach, L. (1951). Coefficient alpha and the internal consistency of tests. Psychometrika, 16, 297–334.
Dong, M., Giles, W. H., Felitti, V. J., Dube, S. R., Williams, J. E., Chapman, D. P., & Anda, R. F. (2004). Insights into causal pathways for ischemic heart disease
adverse childhood experiences study. Circulation, 110(13), 1761–1766.
Dube, S. R., Anda, R. F., Felitti, V. J., Chapman, D. P., Williamson, D. F., & Giles, W. H. (2001). Childhood abuse, household dysfunction, and the risk of
attempted suicide throughout the life span: Findings from the Adverse Childhood Experiences Study. Journal of the American Medical Association,
286(24), 3089–3096.
Dube, S. R., Williamson, D. F., Thompson, T., Felitti, V. J., & Anda, R. F. (2004). Assessing the reliability of retrospective reports of adverse childhood
experiences among adult HMO members attending a primary care clinic. Child Abuse & Neglect, 28(7), 729–737.
Dube, S. R., Miller, J. W., Brown, D. W., Giles, W. H., Felitti, V. J., Dong, M., & Anda, R. F. (2006). Adverse childhood experiences and the association with ever
using alcohol and initiating alcohol use during adolescence. Journal of Adolescent Health, 38(4), 444–e1.
Edwards, V. J., Holden, G. W., Felitti, V. J., & Anda, R. F. (2003). Relationship between multiple forms of childhood maltreatment and adult mental health in
community respondents: Results from the adverse childhood experiences study. American Journal of Psychiatry, 160(8), 1453–1460.
Feder, A., Nestler, E. J., & Charney, D. S. (2009). Psychobiology and molecular genetics of resilience. Nature Reviews Neuroscience, 10(6), 446–457.
Finkelhor, D., Ormond, R., Turner, H., & Hamby, S. L. (2005). The victimization of children and youth: A comprehensive national survey. Child Maltreatment,
10, 5–25.
Ford, E. S., Anda, R. F., Edwards, V. J., Perry, G. S., Zhao, G., Li, C., & Croft, J. B. (2011). Adverse childhood experiences and smoking status in five states.
Preventive Medicine, 53(3), 188–193.
George, D., & Mallery, P. (2003). SPSS for Windows step by step: A simple guide and reference, 11.0 update (4th ed.). Boston: Allyn & Bacon.
Green, C. A., & Pope, C. R. (2000). Depressive symptoms, health promotion, and health risk behaviors. American Journal of Health Promotion, 15(1), 29–34.
Hardt, J., & Rutter, M. (2004). Validity of adult retrospective reports of adverse childhood experiences: Review of the evidence. Journal of Child Psychology
and Psychiatry, 45(2), 260–273.
Heim, C., Newport, D. J., Mletzko, T., Miller, A. H., & Nemeroff, C. B. (2008). The link between childhood trauma and depression: Insights from HPA axis
studies in humans. Psychoneuroendocrinology, 33(6), 693–710.
Kendler, K. S., Bulik, C., Silberg, J., Hettema, J., Myers, J., & Prescott, C. (2000). Childhood sexual abuse and adult psychiatric and substance use disorders in
women: An epidemiological and co-twin control analysis. Archives of General Psychiatry, 57, 953–959.
Kessler, R. C., Berglund, P., Demler, O., Jin, R., Koretz, D., Merikangas, K. R., . . . & Wang, P. S. (2003). The epidemiology of major depressive disorder: Results
from the National Comorbidity Survey Replication (NCS-R). Journal of the American Medical Association, 289(23), 3095–3105.
Kroenke, K., Spitzer, R. L., & Williams, J. B. (2001). The PHQ-9: Validity of a brief depression severity measure. Journal of General Internal Medicine, 16(9),
606–613.
Löwe, B., Spitzer, R. L., Gräfe, K., Kroenke, K., Quenter, A., Zipfel, S., . . . & Herzog, W. (2004). Comparative validity of three screening questionnaires for
DSM-IV depressive disorders and physicians’ diagnoses. Journal of Affective Disorders, 78(2), 131–140.
100 J.C. Poole et al. / Child Abuse & Neglect 64 (2017) 89–100

Luthar, S. S., & Cicchetti, D. (2000). The construct of resilience: Implications for interventions and social policies. Development & Psychopathology, 12(4),
857–885.
Lynskey, M. T., & Fergusson, D. M. (1997). Factors protecting against the development of adjustment difficulties in young adults exposed to childhood
sexual abuse. Child Abuse & Neglect, 21, 1177–1190.
Mathers, C. D., & Loncar, D. (2006). Projections of global mortality and burden of disease from 2002 to 2030. PLoS Medicine, 3(11), e442.
Montpetit, M. A., Bergeman, C. S., Deboeck, P. R., Tiberio, S. S., & Boker, S. M. (2010). Resilience-as-process: Negative affect, stress, and coupled dynamical
systems. Psychology and Aging, 25(3), 631–640.
Morris, M. C., Kouros, C. D., Fox, K. R., Rao, U., & Garber, J. (2014). Interactive models of depression vulnerability: The role of childhood trauma,
dysfunctional attitudes, and coping. British Journal of Clinical Psychology, 53(2), 245–263.
Nelson, E. C., Heath, A. C., Madden, P. A., Cooper, M. L., Dinwiddie, S. H., Bucholz, K. K., . . . & Martin, N. G. (2002). Association between self-reported
childhood sexual abuse and adverse psychosocial outcomes: Results from a twin study. Archives of General Psychiatry, 59(2), 139–145.
Ong, A. D., Bergeman, C. S., Bisconti, T. L., & Wallace, K. A. (2006). Psychological resilience, positive emotions, and successful adaptation to stress in later
life. Journal of Personality and Social Psychology, 91(4), 730–749.
Rose, D. T., & Abramson, L. Y. (1992). Developmental predictors of depressive cognitive style: Research and theory. In D. Cicchetti, & S. Toth (Eds.),
Rochester symposium of developmental psychopathology (Vol. IV). Rochester, NY: University of Rochester Press, p. 323–349.
Rutter, M. (2006). The promotion of resilience in the face of adversity. In A. Clarke-Stewart, & J. Dunn (Eds.), Families count. Effects on child and adolescent
development. Cambridge: Cambridge University Press.
Seok, J. H., Lee, K. U., Kim, W., Lee, S. H., Kang, E. H., Ham, B. J., . . . & Chae, J. H. (2012). Impact of early-life stress and resilience on patients with major
depressive disorder. Yonsei Medical Journal, 53(6), 1093–1098.
Sim, J., & Wright, C. C. (2005). The kappa statistic in reliability studies: use, interpretation, and sample size requirements. Physical Therapy, 85(3), 257–268.
Simon, G. E. (2003). Social and economic burden of mood disorders. Biological Psychiatry, 54, 70–75.
Southwick, S. M., Vythilingam, M., & Charney, D. S. (2005). The psychobiology of depression and resilience to stress: Implications for prevention and
treatment. Annual Review of Clinical Psychology, 1, 255–291.
Spitzer, R., Kroenke, K., & Williams, J. (1999). Validation and utility of a self-report version of PRIME-MD: The PHQ primary care study. Journal of the
American Medical Association, 282, 1737–1744.
Steinhardt, M., & Dolbier, C. (2008). Evaluation of a resilience intervention to enhance coping strategies and protective factors and decrease
symptomatology. Journal of American College Health, 56(4), 445–453.
Straus, M. A. (1979). Measuring intrafamily conflict and violence: The conflict tactics (CT) scales. Journal of Marriage and the Family, 75–88.
Sullivan, G. M., & Feinn, R. (2012). Using effect size-or why the P value is not enough. Journal of Graduate Medical Education, 4(3), 279–282.
Wagnild, G. (2003). Resilience and successful aging. Comparison among low and high income older adults. Journal of Gerontological Nursing, 29(12), 42–49.
Wang, P. S., Simon, G., & Kessler, R. C. (2006). The economic burden of depression and the cost effectiveness of treatment. International Journal of Methods
in Psychiatric Research, 12, 22–23.
Werner, E. E., & Smith, R. S. (2001). Journeys from childhood to midlife: Risk, resilience and recovery. Ithaca, NY: Cornell University Press.
Williams, J. W., Noel, P. H., Cordes, J. A., Ramirez, G., & Pignone, M. (2002). Is this patient clinically depressed? Journal of the American Medical Association,
287(9), 1160–1170.
Wingo, A. P., Wrenn, G., Pelletier, T., Gutman, A. R., Bradley, B., & Ressler, K. J. (2010). Moderating effects of resilience on depression in individuals with a
history of childhood abuse or trauma exposure. Journal of Affective Disorders, 126(3), 411–414.
Wright, M. O., Masten, M. S., & Narayan, A. J. (2013). Resilience processes in development: four waves of research on positive adaptation in the context of
adversity. In R. B. Brooks (Ed.), Handbook of resilience in children (pp. 15–37). New York, NY: Springer Science+Business Media.
Wyatt, G. E. (1985). The sexual abuse of Afro-American and white-American women in childhood. Child Abuse and Neglect, 9(4), 507–519.